One$-$Hot Encoding $(20$ Points$)$

TO$-$DO

It is assumed that a dataset that has a class attribute $($used for a classification problem$)$ has $10$ classes

that are recorded as numeric values: $0,1,2,..9.$ However, these numeric values are assigned to the

classes arbitrarily. There is no ordinal ranking meaning among them. It is similar to the case of the

values $0$ and $1$ that are used to represent NO $($or NEGATIVE$)$ and YES $($or POSITIVE$).$

The dataset will be used in a deep learning project.

Question $1\mathrm{.}1$:

Using critical thinking and based on the lectures, is it necessary to perform any kind of coding $($integer

coding or one$-$hot coding$)$ on these class values for better performance of the project?

Question $1\mathrm{.}2$:

If the answer to Question $1\mathrm{.}1$ is $$YES$,$ what type$($s$)$ of encoding needs to be done to process the class

values before using the dataset for the deep learning project?

Question $1\mathrm{.}3$:

Based on the answer to Question $1\mathrm{.}2,$ explain the steps of what needs to be done for each type of

encoding.

Question $1\mathrm{.}4$:

Based on the answer to Question $1\mathrm{.}2,$ perform the necessary encoding tasks to transform the class values

before using the dataset for the deep learning project.

SUBMISSION REQUIREMENT #$1$

$--)$ Provide the answers$/$solutions to each of the above questions.

$--)$ Display the results of the encoding task.

$--)$ All contents are saved in an MS DOCS document named $$Assignment $3$ MLPs on Iris with

Keras